Earth faces a serious threat from asteroids, and planetary security depends on knowing their paths and their dangers. The need to use data and machine learning to more accurately forecast and evaluate these hazards served as the impetus for this effort. Cosmic Collision analyzes data of asteroids using a machine learning model to forecast possible collision hazards and offer insightful information to lessen dangers. Based on in-depth data analysis, it displays the sizes, trajectories, and impact probability of asteroids. Python and a number of data science libraries, such as pandas, NumPy, and scikit-learn, were used in the project's construction. I created an algorithm that can recognize patterns and predict possible collisions by training the model using information from NASA and other sources. Organizing the large and intricate databases on asteroid features and orbits was the most difficult task. A significant amount of work and computer resources were needed to process and refine this data in order to produce a useful, predictive model. I take pride in developing a useful and precise model that can forecast the likelihood of asteroid collisions. Another significant accomplishment was presenting asteroid data in an understandable and useful manner, which allowed a larger audience to access the insights. I gained extensive knowledge in feature engineering, data preparation, and the complexities of astronomical datasets. I also improved my knowledge of machine learning methods that are appropriate for predictive modeling in high-stakes situations, such as asteroid risk assessment. Next, by adding real-time data and improving our algorithms, I hope to increase the model's accuracy. In order to enhance data accessibility and incorporate the model into larger planetary defense projects, I also hope to work with space research groups.
Built With
- dataset
- machine-learning
- matplotlib
- numpy
- pandas
- python